基于时间窗约束的多人多储位拣选路径优化模型及改进遗传算法应用研究

Optimal model of multi-person and multi-location picking path based on time window constraint and application of improved genetic algorithm

针对目前主要的单人单储位拣选路径规划研究现状,以安徽BY零件库为研究对象,考虑实际存在的多人同时拣选与多储位问题(MLP),建立多储位下的多人拣选路径优化模型。首先确定拣选的人数;然后针对多人拣选路径优化模型提出一种基于时间窗约束的改进遗传算法(TWC-IGA)进行模型的求解,由IGA优化路径并得到拣货路径集合,再利用时间窗约束和两阶段策略进行路径的冲突预测、避障与动态局部调整;最后,通过仿真实验与遗传算法、贪心算法进行对比,验证文中所提的TWC-IGA的稳定性与收敛速度更好,结果更优,极大地提高拣选效率,对实际拣选具有有效性。

Taking Anhui BY's parts library as the research object,Aiming at the current research status of single-person and single-storage picking path planning,the actual existing multi-person simultaneous picking and Multi-Location Problem(MLP)was considered,and the multiple storage and multi-person picking path optimization model was established.The number of picking people was first determined.For the multi-person picking path optimization model,an Improved Genetic Algorithm for Time Window Constraints(TWC-IGA)was proposed to solve the model.The path was optimized by IGA and the picking route collection was obtained.Then the time window constraints and two-stage strategy was used for route conflict prediction,obstacle avoidance and dynamic local adjustment.Through the simulation experiments,the proposed model was compared with genetic algorithms and greedy algorithms.The result showed that the stability and convergence speed of the proposed model were better,and the picking efficiency was greatly improved,which was effective for actual picking.